Too Hot?

[River]

I just got a i9 14900k and use a AIO water cooler my temperatures are 37 C idle around and mainly 60 to 70 C when gaming on some games. Games that use the CPU more raise the temperature to max 78 C give or take.

[r.linder]

Yeah that's absolute nonsense. The max temp listed by Intel and AMD is the maximum operating temperature, when you're running at factory settings, any value below the maximum operating temperature is considered within specification and not an immediate danger to the processor, it's designed to NOT KILL ITSELF. How hard is that to understand?

AMD has also plainly stated that it's normal for their chips to run up to 95 degrees and that it really won't harm the processor.

[Bad 💀 Motha]

I do agree in that I simply don't want anything running that hot and I will try my best to keep it running cooler if I can. If you run multiple PCs all day within a single room you'll understand where I'm coming from on that. The room literally heats up after a couple hours, it's ridiculous. On the upside I don't need any extra heating in that room in the winter time.

[r.linder]

Bad 💀 Motha님이 먼저 게시:

I do agree in that I simply don't want anything running that hot and I will try my best to keep it running cooler if I can. If you run multiple PCs all day within a single room you'll understand where I'm coming from on that. The room literally heats up after a couple hours, it's ridiculous. On the upside I don't need any extra heating in that room in the winter time.

Agreed, I care more about noise than the actual thermals though

[🦊Λℚ𝓤ΛƑΛᗯҜᔕ🦊]

Angry Gandalf님이 먼저 게시:

gamer님이 먼저 게시:

I just got a i9 14900k and use a AIO water cooler my temperatures are 37 C idle around and mainly 60 to 70 C when gaming on some games. Games that use the CPU more raise the temperature to max 78 C give or take.

60-70 is WAY too hot for a CPU and it might be possible, that you're destroying your CPU RIGHT NOW!!!!!!
I'm glad you've decided to start a thread instead of using Google for 5 seconds.

First things first.
Please remove the cooler from your CPU and attach the custom boxed cooler.
When you're ready, hit me up.

This is not even true at all. OP: I hope you don't listen to comments like this. That's complete false information.

The maximum temperature for AMD Ryzen CPU's is 95c, Intel CPU's is 105c. Anything below that is perfectly fine and nothing to be concerned about.

[Bad 💀 Motha]

Intel anything is basically fine up to around 100*C

Ryzen 7700X/7800X3D should not run above 90*C

Ryzen 7900X/7950X should not run above 100*C ~ These can easily operate under high or full loads all day long @ 95*C without any issues.

[Guydodge]

🦊Λℚ𝓤ΛƑΛᗯҜᔕ🦊님이 먼저 게시:

Angry Gandalf님이 먼저 게시:

60-70 is WAY too hot for a CPU and it might be possible, that you're destroying your CPU RIGHT NOW!!!!!!
I'm glad you've decided to start a thread instead of using Google for 5 seconds.

First things first.
Please remove the cooler from your CPU and attach the custom boxed cooler.
When you're ready, hit me up.

This is not even true at all. OP: I hope you don't listen to comments like this. That's complete false information.

The maximum temperature for AMD Ryzen CPU's is 95c, Intel CPU's is 105c. Anything below that is perfectly fine and nothing to be concerned about.

while your correct everything this guy says is fake news.on the same note i personally
do not believe intel or amd that these new basically overclocked super chips are
intended to run at 95-100c that is to hot.they say they run that hot to hit the numbers
so out of box users are not alarmed and honestly they should be.
claiming its "normal"doesnt mean degradation of the cpu wont happen quicker at these
temps.do not kid yourself theres things you should do to keep these chips running
at there peak performance for a extended period.if you run these chip at 95--100c
it is a problem.they can an will run cooler without any lose in performance.
these are not plug and play chips you need to learn to correctly fine tune them.

the degradation will take years to happen and most who initially buy these
chips wont experience it as we tend to rebuild before it would become a
noticeable drop.

my goal is to run 75c or lower under my normal running conditions i could
care less what it does under bench marks.im not running bench marks
im running games.people are to hung up on bench marks i dont bench
mark my builds ever.i use real world tests.im not going to sacrifice performance
because of the unreal running conditions of a bench mark.if you using
applications that actually run the chip that hard then absolutely use them
but for the vast majority its unrealistic and unnecessary.
i use cpu-z only to make sure my cpu and memory are performing normally.
as you can actually lose power on both overclocking to much.
i find cpu-z is very close to a heavy gaming load and works great for that.
just run bench CPU and then submit and compare it will tell you everything you need to know.

[MonkehMaster]

6950x here and had a few 6850k's before that, they never went over 45c unless my room ac couldnt keep the room cool enough (70f-75f), thats while playing games, runs much lower on boot and idle.

no its not to hot, but cooler is always better.

few things to consider.

having your radiators set so they are blowing out of your case, not inside it.

setting aggressive fan curves and keep the temps in the pc room cool enough.

changing the thermal paste when needed and keeping the pc, vent, filters, ect.. free from dust.

make sure all your fans are working and facing the right way.

also to mention, protection limits while said limits may throttle to lower heat, its not 100% safe from over heating and killing itself and yes they can kill themselves despite people thinking they cant (i have seen both cpu's and gpu's kill themselves with heat and not work again afterwards), so if you want your hardware to last longer and run like new, your best bet is to maintain said stuff by cleaning, re-pasting and keeping them cooler.

[C1REX]

I think people focus too much on the lowest temps when hot spot and junction temps should get more attention.

General CPU and GPU temps may show 75C and hide serious cooling problems. I don’t understand why hot spot, junction or in general the highest temp is not shown in the most basic monitors. Like really? Who cares about the lowest temp? It only makes the temp delta between hottest part bigger and worse.

[r.linder]

C1REX님이 먼저 게시:

I think people focus too much on the lowest temps when hot spot and junction temps should get more attention.

General CPU and GPU temps may show 75C and hide serious cooling problems. I don’t understand why hot spot, junction or in general the highest temp is not shown in the most basic monitors. Like really? Who cares about the lowest temp? It only makes the temp delta between hottest part bigger and worse.

If hotspot was that big of a deal then the hotspot maximum wouldn't be higher than the maximum temperature for the entire core. AMD's maximum temperature (hotspot) for RX 5000 through 7000 series is 110 degrees (C), NVIDIA's is 104.9 degrees (C). Hotspot is not a metric even used on the CPU and the CPU is still designed to run up to its maximum operating temperature before throttling.

Don't forget that professionals are engineering these processors and they're designed to run a certain way out of the box that won't just burn themselves out, and don't forget that unless you're actually working for Intel, AMD, etc. on engineering these parts, you're in no way qualified to suggest that the way the components are designed to operate is unsafe or wrong. It's not like the parts won't detect the hottest part of the core, that isn't how it works, it's all handled automatically in firmware out of the box.

[C1REX]

尺．し工几ᗪヨ尺님이 먼저 게시:

Don't forget that professionals are engineering these processors and they're designed to run a certain way out of the box that won't just burn themselves out, and don't forget that unless you're actually working for Intel, AMD, etc. on engineering these parts, you're in no way qualified to suggest that the way the components are designed to operate is unsafe or wrong. It's not like the parts won't detect the hottest part of the core, that isn't how it works, it's all handled automatically in firmware out of the box.

I don't argue with that.
I simply would prefer that even the simplest temperature monitoring software (MSI Afterburner for example) would show the highest temperature on the CPU and GPU.
Especially GPU can be misleading. People had GPUs (7900XTX) running below 70C while thermal throttling as the junction temp was hitting 105C. Many metrics can be very misleading if not checked properly.

Small temp difference and big power consumption difference.
https://i.ibb.co/q5b3CxJ/CPU-temp-and-power.jpg

Very big temp difference.
https://i.ibb.co/BjpNGp9/GPU-temp.jpg


The data is correct but can be misleading.
CPUs can be heavily thermal throttled by motherboards hot VRM.

[MonkehMaster]

C1REX님이 먼저 게시:

尺．し工几ᗪヨ尺님이 먼저 게시:

Don't forget that professionals are engineering these processors and they're designed to run a certain way out of the box that won't just burn themselves out, and don't forget that unless you're actually working for Intel, AMD, etc. on engineering these parts, you're in no way qualified to suggest that the way the components are designed to operate is unsafe or wrong. It's not like the parts won't detect the hottest part of the core, that isn't how it works, it's all handled automatically in firmware out of the box.

I don't argue with that.
I simply would prefer that even the simplest temperature monitoring software (MSI Afterburner for example) would show the highest temperature on the CPU and GPU.
Especially GPU can be misleading. People had GPUs (7900XTX) running below 70C while thermal throttling as the junction temp was hitting 105C. Many metrics can be very misleading if not checked properly.

Small temp difference and big power consumption difference.
https://i.ibb.co/q5b3CxJ/CPU-temp-and-power.jpg

Very big temp difference.
https://i.ibb.co/BjpNGp9/GPU-temp.jpg


The data is correct but can be misleading.
CPUs can be heavily thermal throttled by motherboards hot VRM.

i run majority of my fans according to vrm temps (7 fans... 3 upper, 1 rear, 1 bottom, 2 front of case), cpu fans (2 with heatsink) on cpu temps, with my gpu fans (triple fan card) and 2 case fans (2 fans mounted directly across storage rack on radiator bracket, facing across and below the gpu) on the 2 extra fan headers on the gpu running at gpu temps (blows across my pcie add-in card for 2 m.2's and a capture card, plus airflow into the gpu fans), in any case, my jet engine pc makes loads of noise but stays as cool as the room its in.

anywho, dont most monitoring programs show those temps? i agree though as those temps should be part of those tests/ect... and i also never see vrm temps as much of a discussion, those things run much hotter and also tend not to get much airflow.

[Illusion of Progress]

C1REX님이 먼저 게시:

I don't argue with that.
I simply would prefer that even the simplest temperature monitoring software (MSI Afterburner for example) would show the highest temperature on the CPU and GPU.

It seems more hidden on GeForce in Afterburner for... some reason (?) but Radeon certainly shows this. Or maybe newer GeForces show it now too so I thought maybe it's an age thing but most other software shows it. Afterburner just doesn't.

https://i.imgur.com/zAmmuTm.png

https://i.imgur.com/lu7Tbgk.png

[Shaggin'Wagon]

MonkehMaster님이 먼저 게시:

i run majority of my fans according to vrm temps (7 fans... 3 upper, 1 rear, 1 bottom, 2 front of case), cpu fans (2 with heatsink) on cpu temps, with my gpu fans (triple fan card) and 2 case fans (2 fans mounted directly across storage rack on radiator bracket, facing across and below the gpu) on the 2 extra fan headers on the gpu running at gpu temps (blows across my pcie add-in card for 2 m.2's and a capture card, plus airflow into the gpu fans), in any case, my jet engine pc makes loads of noise but stays as cool as the room its in.

anywho, dont most monitoring programs show those temps? i agree though as those temps should be part of those tests/ect... and i also never see vrm temps as much of a discussion, those things run much hotter and also tend not to get much airflow.

You should be aware that by connecting fans to fan headers on your video card then the power draw of those fans will be factored in to the video card's power limit. Which means the fans can cause your video card to trigger it's power limit and throttle performance when the fans are running like that. It's usually a bad idea to do that and it should be avoided. You want your video card it's self to use all of it's power limit, not losing performance due to fans connected to it causing artificial power limit triggering.

[MonkehMaster]

Shaggin'Wagon님이 먼저 게시:

MonkehMaster님이 먼저 게시:

i run majority of my fans according to vrm temps (7 fans... 3 upper, 1 rear, 1 bottom, 2 front of case), cpu fans (2 with heatsink) on cpu temps, with my gpu fans (triple fan card) and 2 case fans (2 fans mounted directly across storage rack on radiator bracket, facing across and below the gpu) on the 2 extra fan headers on the gpu running at gpu temps (blows across my pcie add-in card for 2 m.2's and a capture card, plus airflow into the gpu fans), in any case, my jet engine pc makes loads of noise but stays as cool as the room its in.

anywho, dont most monitoring programs show those temps? i agree though as those temps should be part of those tests/ect... and i also never see vrm temps as much of a discussion, those things run much hotter and also tend not to get much airflow.

You should be aware that by connecting fans to fan headers on your video card then the power draw of those fans will be factored in to the video card's power limit. Which means the fans can cause your video card to trigger it's power limit and throttle performance when the fans are running like that. It's usually a bad idea to do that and it should be avoided. You want your video card it's self to use all of it's power limit, not losing performance due to fans connected to it causing artificial power limit triggering.

havent had any issues with them on it yet and the pc has been in use for 7-8 years.

[Shaggin'Wagon]

MonkehMaster님이 먼저 게시:

Shaggin'Wagon님이 먼저 게시:

You should be aware that by connecting fans to fan headers on your video card then the power draw of those fans will be factored in to the video card's power limit. Which means the fans can cause your video card to trigger it's power limit and throttle performance when the fans are running like that. It's usually a bad idea to do that and it should be avoided. You want your video card it's self to use all of it's power limit, not losing performance due to fans connected to it causing artificial power limit triggering.

havent had any issues with them on it yet and the pc has been in use for 7-8 years.

I don't think you were understanding what I was trying to say, maybe I didn't explain it correctly. I will try again: Video cards have a power limit. When that limit is reached the video card will reduce the clock speed of the core and throttle it. If you power fans from the video card then the fans are taking a % of your video card's power limit away that could be used for rendering graphics. It would appear that everything is working fine but you are losing some performance by doing that. Your video card would run faster if it didn't have fans connected to it.

For example: If you have a video card with a 300 watt power limit and connected two 20-watt fans to it then you would be reducing performance by -14%, worse if you connected more powerful fans to it.